Multiple Cooling Tower Control Strategies in IESVE

Date Published

2nd Apr 2020

Birajan Bhandari
Building Performance Engineer, IES

This technical article compares the predicted energy consumption of different cooling tower control strategies when the building cooling load is known, and the chiller(s) design has been established. 

The available heat-rejection devices in ApacheHVAC include a closed-circuit fluid coolers or an open-circuit cooling towers. However, the closed-circuit towers are not used as often1, due to higher cost and reduced efficiency owing to the added approach of a heat exchanger, this comparison will focus on open circuit cooling tower. In commercial buildings, most of the waste heat, being rejected via a cooling tower, comes from the condenser side of the chiller(s).

Figure 1: Multiple cooling tower control strategies 

Cooling Tower Control Strategies:

In IESVE Software (Version 2019 FP2) three different types of multiple cooling tower operation/control strategies are compared. In all cases, the chillers are rejecting the same amount of heat to the CW Loop.

  1. Maximize cell operation: This control strategy will operate as many user-defined cells as possible regardless of how many chillers are operating, subject to the constraint of minimum pump flow fraction and fan operation fraction. 
  2. Interlocked with chillers: In this type of control strategy, the number of cooling towers is equal to the number of chillers, which is a requirement of some performance-based energy codes E.g. Title-24 2019 Standard Design Model. During operation, the Nth cooling tower turns on only when the associated chiller is rejecting heat (1-to-1 operation).
  3. N+1 cooling tower per chiller: This strategy allows an identical number of cooling towers to chillers, or with one additional cooling tower (N+1). In Figure 1., a design with 4 chillers will allow 4 (N) or 5 (N+1) cooling towers. During operation, two cooling towers turn on when one chiller is operational, and all five cooling towers turned on when four chillers were operational (between 3:30 pm to 6:00 pm in Figure 2).  

Note: The energy codes (ASHRAE 90.1, IECC, and Title 24) require the use of maximum number of cells as possible. This can be modeled in ApacheHVAC using the “Maximize cell operation” control strategy. 2

Comparison of Cooling Tower Fan Energy:

The total fan energy for each of the cooling towers design options are compared in Figure 2. For this predictive analysis, all options assume that the cooling tower fans are on VSD control, minimum fan flow fraction of 0.20 and the approach temperature of the cooling towers is 10°F. In this particular design scenario, the most efficient control strategy for cooling tower fan energy is the “Maximize cell operation” control strategy with 5 cooling towers, followed by “N+1 Cooling tower per chiller” control strategy with 5 cooling towers. The least efficient control strategy is “Interlocked with chillers” which has four cooling towers.

Figure 2: Fan energy Consumption of the three cooling tower Operation/control strategies 

Comparison of Cooling Tower Airflow:

The total fan airflow of each cooling towers is compared in Figure 3. The cooling tower airflow is calculated by APACHE and are shown in the VistaPro. The cooling tower fan airflow follows the same hourly trend as the cooling tower heat rejection plot.

Figure 3: Cooling tower Fan Airflow 

Conclusions

  1. “N+1 Cooling Tower per Chiller” and “Interlocked with Chiller” control strategies are dependent on the number of chiller(s) that are operating. For this design example, the chiller design had already incorporated an N+1 design, so sequenced staging of chiller operation is important. 
  2. Based on the predictive energy modeling, the “Maximize Cell Operation” is the most efficient cooling tower strategy method for this particular design in this particular climate of San Jose, CA. 
  3. One of the most important cooling tower variables is the approach temperature. The “Approach” is defined as the difference between the cold-water temperature and either the ambient or entering wet-bulb temperature. A smaller approach, which may result in an over-sized/expensive cooling tower, will provide lower cooling tower leaving temperature. This, in turn, will result in lower fan energy and higher chiller efficiency due to reduced lift on the chiller. Modern cooling towers commonly have an approach temperature as low as 5 °F. 
  4. Annual climatic data (WBT) and design weather (WBT) are important when considering evaporative coupling at a constant 5 °F approach. This is due to the enthalpic difference of air at the 5°F approach under different design WBT.  VistaPro includes a number of methods, including custom variables, to evaluate evaporative cooling, including wet bulb depression at 5°F, 10°F, etc (Figure 4).

Figure 4: Wet-Bulb Depression at 5°F, comparing San Francisco and Sacramento  

 

 Cover photo courtesy of Baltimore Aircoil Company


1 Taylor, Steven T. “Optimizing Design & Control of Chilled Water Plants” ASHRAE Journal, Mar. 2012, pp. 60-66

2 Morrison Frank “Saving Energy with Cooling Towers” ASHRAE Journal, Feb. 2014, pp. 34-40

Latest from DiscoverIES